Managed Gateway Quick Start
Get your first AI request flowing through Axemere's Managed Gateway in about 5 minutes: no infrastructure to deploy.
Table of Contents
- Prerequisites
- Step 1: Activate a gateway
- Step 2: Enable a provider and add a credential
- Step 3: Create a gateway key
- Step 4: Send your first request
- Next Steps
Prerequisites
- An Axemere account: sign up at console.axemere.ai
- An API key for the AI provider you want to use (OpenAI, Anthropic, etc.)
Step 1: Activate a gateway
Open Services in the console. One Managed Gateway package is available: Growth Pack, purchased in quantity as usage grows, ideal for individual developers and production teams alike.
Growth Pack includes a 30-day free trial: no credit card required.
Choose a quantity (1, 5, 10, or 25 packs, or a custom number) and activate. Once activated, a gateway endpoint is provisioned for your org:
https://us.gw.axemere.ai
Step 2: Enable a provider and add a credential
2a. Enable a provider: Open Providers and enable the provider you want to use (for example, Anthropic or OpenAI).
2b. Add a credential: Open Credentials and add a credential for that provider. This is your AI provider API key. The gateway uses it to authenticate on your behalf, so your developers never need to handle AI provider keys directly. Rotating a compromised or expired key is a one-step operation in the console with no application-side changes required.
Step 3: Create a gateway key
Open Gateway Keys and create a new gateway key. This is the key your applications and tools use to authenticate to the Axemere Managed Gateway; it is not an AI provider key.
Warning: Copy the key now; it is only shown once. If you lose it, you'll need to create a new one.
Step 4: Send your first request
Set the gateway URL and your gateway key, then send a request:
export AXEMERE_KEY="mvgc_k_your-key-here"
export GW="https://us.gw.axemere.ai"
curl -s "$GW/proxy/anthropic/k/$AXEMERE_KEY/v1/messages" \
-H "Content-Type: application/json" \
-H "anthropic-version: 2023-06-01" \
-d '{
"model": "claude-haiku-4-5",
"max_tokens": 64,
"messages": [{"role": "user", "content": "Hello"}]
}' | jq .A successful response returns the provider's API response. Check Records in the console. Your request should appear within a few seconds.
The URL structure:
https://us.gw.axemere.ai/proxy/{provider}/k/{axemere-key}
/proxy/anthropicor/proxy/openai: routes to that provider's API (Gemini, Azure OpenAI, and any provider with an OpenAI-compatible chat completions API also use the/proxy/openaipath)/k/mvgc_k_...: your Axemere gateway key (authenticates you to the managed gateway)
Next Steps
| Task | Where to go |
|---|---|
| Route VS Code or Claude CLI through the gateway | AI Tools |
| Configure policies and rate limits | Configure |
| View request records and spend | Records |
| Add team members | Team |
| Set up workloads for attribution | Workloads |